Refactor internal engine #27082
Refactor internal engine #27082
Conversation
| } | ||
|
|
||
| @FunctionalInterface | ||
| protected interface SeqNoServiceSupplier { |
There was a problem hiding this comment.
why not use java.util.function.BiFunction?
There was a problem hiding this comment.
This is a good catch. This is a product of the fact that I initially started with the supplier being a function of more parameters (the shard ID, the allocation ID, etc.) but collapsed these into the two that are left and did not see until you pointed it out that we can go all the way and remove the interface. Thank you.
| @@ -2008,8 +2031,7 @@ public MergeStats getMergeStats() { | |||
| return mergeScheduler.stats(); | |||
| } | |||
|
|
|||
| @Override | |||
| public SequenceNumbersService seqNoService() { | |||
| public final SequenceNumbersService seqNoService() { | |||
There was a problem hiding this comment.
Just wondering: Is it possible to move seqNoService field to Engine make the seqNoService() method there final? The SeqNoServiceSupplier (or BiFunction) should then also be moved as an constructor parameter to Engine.
There was a problem hiding this comment.
I don't think we can do this cleanly right now. It means that the sequence number service has to be available immediately (to pass to the super constructor) but we are also trying to initialize the index writer at the same time that we initialize the sequence number service.
There was a problem hiding this comment.
I see, that makes it trickier. Thanks for explaining that.
| import static java.util.Collections.emptyMap; | ||
| import static org.elasticsearch.index.mapper.SourceToParse.source; | ||
|
|
||
| public class TranslogHandler implements EngineConfig.TranslogRecoveryRunner { |
There was a problem hiding this comment.
nit: maybe add class jdoc here about what the purpose of this implementation is?
|
@s1monw I'm happy to pull this into master/6.x rather than a long-running feature branch, it will require maintenance on the feature branch then. 😇 |
This commit is a minor refactoring of internal engine to move hooks for generating sequence numbers into the engine itself. As such, we refactor tests that relied on this hook to use the new hook, and remove the hook from the sequence number service itself.
jasontedor
force-pushed
the
following-engine
branch
from
5fb2f0f to
c55542c
Compare
This commit is a minor refactoring of internal engine to move hooks for generating sequence numbers into the engine itself. As such, we refactor tests that relied on this hook to use the new hook, and remove the hook from the sequence number service itself. Relates #27082
clintongormley
added
:Distributed Indexing/Distributed
This commit is a minor refactoring of internal engine to move hooks for generating sequence numbers into the engine itself. As such, we refactor tests that relied on this hook to use the new hook, and remove the hook from the sequence number service itself.